The present invention relates to a main station and a substation in a network, e.g. in a field bus network, and in particular the reception of external data and for the transmission of device-internally prepared data via the network. The invention also relates to a method for transmitting data in a network.
Data networks, for example a field bus network, usually have a main station and one or more substations that are connected to one another by means of point-to-point connections. The main station and the substation constitute a ring network in which the main station is connected via a unidirectional network connection to a data input of the first substation, a data output of the first substation is connected to a data input of a second substation, a data output of the second substation is connected to a data input of a third substation, etc. The data output of the last substation is connected to a data input of the main station.
Alternatively, a linear network topology can also be provided in which the main station and the substations are connected to each other via network connections that have two channels for the send and receive directions.
A hallmark of data networks is that the data received in the substations are relayed via a repeater. The time at which the relay occurs in relation to the reception can be delayed as needed and the received data can be entirely or partially prepared for further processing and evaluation in the substation. For example, the data can be transmitted in data packets.
The communication on a data network comprised in this way can occur, for example, randomly, i.e. the communication time at which a station transmits a data packet via its data output to the data network is not fixed ahead of time. For example, each station can send the data packet via its data output at any time. In an IP network protocol (IP=Internet protocol) of this kind, since the substations of the network as a rule transmit data packets arriving at their data input immediately to the data output via the repeater, at the beginning of the transmission of a data packet, overlaps of two data packets can occur if an external data packet is received during the transmission. This external data packet can no longer be interpreted by a subsequent network subscriber.
In data networks, it is also possible to provide two or more communication channels in time domains that are reserved for them. In addition to the above-described IP communication channel, it is also possible to provide a time domain that is referred to as a real-time channel, on which data packets are transmitted in a systematic sequence; a chronologically defined time window in at least one of the data packets is associated with a respective substation so that the substation can read the respective data associated with it from the data packet.
A time window is also reserved for the substation in another data packet and the substation can insert data into this time window while the substation is forwarding the received data packet to its data output.
Particularly in the IP communication channel, the problem arises that two overlapping data packets become unusable and thus hinder data communication on the data network.